"By knowing his business. When he passes on a foundry, for example, he ought to know, first, the fire record of foundries in general; second, what rate of premium they ought in general to pay; and third, what the dangers, or, as we call them, hazards, are. By looking at the map he must be able to tell where the fire is most likely to start--where, in other words, fires usually do start in foundries. Probably it will be the cupola charging platform or the core ovens. Then he can closely tell from the construction of that particular foundry, considering also the protection, extinguishing appliances, public water pressure, nearness of the fire department, and fifty other considerations, how much of the whole plant would burn--probably. If only half, then he feels safe in writing ten thousand dollars on the risk, since only half of it is likely to be destroyed by one fire." "I don't see how you can tell." "Well, most companies have quite elaborate line sheets to assist their underwriters in determining how much to hold on various classes of risks, but between you and me, you _can't_ tell surely. But you do the best you can, and the ablest underwriter is the man who tells the closest. A really good underwriter should know the hazards of all the ordinary risks in the world, and be able to tell you offhand what is the danger point in a brewery, a playing-card factory, a paper mill, a public school, a shovel works, a Catholic church, a chemical laboratory--every sort and kind of risk. Of course he has surveys, made by inspectors, to help him, showing details the map fails to show, such as the location of your piano, and where the hazards lie and how they are cared for. But inspectors are fallible, and he must _know_--everything." "You make my head whirl," Helen said. "To know everything! It sounds colossal. Do you know everything?" Smith laughed. "No," he replied. "Decidedly not. I'm afraid I know only a very small proportion of what I ought. But the big men of the business do. There is one man who I verily believe is perfectly familiar with every kind of risk in the United States. If there is a chemical process he doesn't know or can't find out about, I'll eat the thing myself. He knows every explosive mixture, every fulminate, every sort or manner of dust, paste, or grease which burns or explodes of itself." "But that one man must be a genius! What does the average man do? Doesn't he need some one to help him in all this? It sounds like such a terrific undertaking to keep track of so many things. Doesn't it make your own head swim at times?" "Well," said Smith, "of course there are a thousand and one things in the nature of aids to the underwriter--things whose proper action he doesn't directly control, although he has to keep a father's eye on them to see that they don't run amuck." "Such as what?" asked the girl.